Situational effects on gender role perceptions in women: A look at the validity of the BEM Sex Role Inventory

Validity of the Bem Sex Role Inventory was examined in a descriptive interaction comparisons design study using volunteers from women attending a large women's health issues conference. Four null hypotheses designed to test variance in selection of the BSRI attributes on the basis of sex and situational needs for each of the BSRI classification groups were tested using three versions of the instrument. Participants were instructed to complete one questionnaire to describe themselves, the second to describe men or women in American society, and the third to describe a man or woman in a vignette situation were placed in random order in a packet along with a personal background information sheet and a letter of explanation. Ninety-one completed questionnaires were returned anonymously by mail to the researcher.

Multiple regression of the feminine and masculine scale scores and the personal characteristics indicated that employment status, educational level, and religious preference exerted influence on part of the scale scores. Discriminant function analysis performed following significant MANOVA results for an analysis of the impact of the order in which the questionnaires appeared in the packet on scale scores demonstrated that over 50% and as much as 75% of the variance on the three masculine scales in the second packet could be attributed to the questionnaire's placement. MANOVA results for each of the four hypotheses were nonsignificant leading to acceptance of the null hypothesis in each case.

Comparison of the results with previous studies lead to the conclusion that although validity of the BSRI could not be ruled out, failure of feminine and masculine subjects to discriminate in selection of attributes on the basis of sex and of androgynous and undifferentiated subjects to differentiate on the basis of situational needs did not supply evidence of content and construct validity. The changing perception of sex roles and expansion of nursing research methodology to include phenomenology were considered as impetus for nurses to conduct phenomenological studies of femininity and masculinity with incorporation of the results in existing or new sex role instruments before conducting further nursing research on sex role identification.